Extract PDF Table using form Extractor

Help Studio
, ,
1

Hi Team,

I have multiple pdf files where I have to extract specific text. The text may not be always in same format but there is a specific keyword which enables us to identify what text to extract? Since the formats are not always same ?How can we achieve this?

Here Milestone is the keyword: Sometimes the text is in tabular format sometimes its not.Its mostly under Performance and Miltesone section most of the times ,sometimes its just text description.

Ist PDF Milestone Details

This image shows a table detailing two project milestones with their respective completion dates, brief descriptions of services to be completed, and associated costs. (Captioned by AI)
This image shows a table detailing two project milestones with their respective completion dates, brief descriptions of services to be completed, and associated costs. (Captioned by AI)1498×672 66.7 KB

I need to capture each milestone details , Amount ,hours,Due Date:

Second PDF Milestone Details:

This image depicts a section from a performance and milestone schedule document detailing services completion guidelines, specific milestones, associated costs, hours, and deadlines, with one milestone being due on November 30, 2023. (Captioned by AI)
This image depicts a section from a performance and milestone schedule document detailing services completion guidelines, specific milestones, associated costs, hours, and deadlines, with one milestone being due on November 30, 2023. (Captioned by AI)1678×1300 372 KB

Forum Engagement Daily Reports
2

Hie @dutta.marina if the keyword is fix then you can use String manipulation method and pass the fix letter as a Reference .
as an example -( pdfOutput.Split(“RefrenceKeyword”)(1).tostring)
and if you have to extract data between 2 reference
pdfOutput.Split(“RefrenceKeyword”)(1).tostring.split(“RefrenceSecond”)(0).tostring.trim
change the reference and index position as per you need .
cheers

3

Hi All,

I have PDF files where I have to extract specific information (Milestone DEtails) from Mile Stone Table. I need to extract the milestone information from the section Performance and Milestone section.The Milestone details are in two different formats given below: How can I achieve this using Document understanding of Regex Extractor:

I need to capture Brief description, Amount Due Date
First format of PDF files

The image is a section from a project document outlining the performance and milestone schedule, detailing the first milestone for migration readiness with a cost of USD 37,000, requiring 104 hours, and due by August 30, 2023. (Captioned by AI)
The image is a section from a project document outlining the performance and milestone schedule, detailing the first milestone for migration readiness with a cost of USD 37,000, requiring 104 hours, and due by August 30, 2023. (Captioned by AI)815×475 83.8 KB

Second Format of PDF FILES:

The image shows a project milestone table detailing two milestones, related to Teradata migration and DLT pipeline implementation, with their descriptions, deadlines, and respective costs. (Captioned by AI)
The image shows a project milestone table detailing two milestones, related to Teradata migration and DLT pipeline implementation, with their descriptions, deadlines, and respective costs. (Captioned by AI)1648×682 129 KB

@Anil_G Any help on this?

4

@dutta.marina

first thing for these table type of data regex is not a best choice

but if you still need then we need some actual samples in pdf …where we can try finding any match and use

cheers

5

@Anil_G

What is the best solution for these data .?

6

@dutta.marina

As this is a proper seggregated table …you can try with du model and training on it to extract the table data and you can as well classify before extracting for different formats

cheers

7

@Anil_G

Can we use form extrator for two different templates?

8

@dutta.marina

use classifier and create different templates in form extractor

cheers

9

@Anil_G

Sent you two pdf samples. Can you help on that?